Convergence issue on MgO cell optimization with def2-TZVP basis set

[myyang]

Hi,

I am new to cp2k, and I would like to do a cell optimization on MgO primitive cell with def2-TZVP basis set (to do some comparation with other all-electron calculation). The input file is mainly generated from Multiwfn 3.8 dev, with OUTER_SCF and settings related to smear added by myself. I use a 4*4*4 supercell with gamma point to avoid Cholesky decompose issue. However, SCF convergence suffers from oscillation. Below is my input file.

Could someone give me some advice? I would really appreciate it.

Ming-Yu Yang

[Marcella Iannuzzi]

Hello 

Since the cell is replicated 4x4x4 also the coordinates need to be replicated 4x4x4 with the keyword in the TOPOLOGY section.

• MULTIPLE_UNIT_CELL

Are you sure that you need smearing and G-space mixing? If not it is preferable to use OT instead of diagonalization. 

Regards

marcella